Presentation is loading. Please wait.

Presentation is loading. Please wait.

CDA 3101 Fall 2013 Introduction to Computer Organization The Arithmetic Logic Unit (ALU) and MIPS ALU Support 20 September 2013.

Similar presentations


Presentation on theme: "CDA 3101 Fall 2013 Introduction to Computer Organization The Arithmetic Logic Unit (ALU) and MIPS ALU Support 20 September 2013."— Presentation transcript:

1 CDA 3101 Fall 2013 Introduction to Computer Organization The Arithmetic Logic Unit (ALU) and MIPS ALU Support 20 September 2013

2 Overview Hardware building blocks ALU design ALU implementation 1-bit ALU 32-bit ALU

3 Hardware Building Blocks ALUs are implemented using lower level components (logic gates) Gate (review) –Hardware element that receives a certain number of inputs and produces one output –Can be represented as a truth table or logic equation –Gates in turn are implemented with transistors ALU Building Blocks (review) –And gate –Or gate –Inverter (not gate) –Multiplexor (mux)

4 Basic Gates

5 Modular ALU Design Facts –Building blocks work with individual (I/O) bits –ALU works with 32-bit registers –ALU performs a variety of tasks (+, -, *, /, shift, etc) Principles –Build 32 separate 1-bit ALUs –Build separate hardware blocks for each task –Perform all operations in parallel –Use a mux to choose the actual operation (make decision) Advantages –Easy to add new operations (instructions) Add new data lines into the muxes; inform “control” of the change.

6 ALU Implementation 1.32-bit ALU uses 32 muxes ( one for each output bit ) 2.Go through instruction set and add data (and control) lines to implement the corresponding operations. Control lines (n) Output: one per mux Data lines (2 n )

7 One-Bit Logical Instructions Map directly onto hardware components –AND instruction One of data lines should be a simple AND gate –OR instruction Another data line should be a simple OR gate A B C 0 1 Op (control)Definition OpC 0A and B 1A or B

8 One-Bit Full Adder Each bit of addition has –Three input bits: A i, B i, CarryIn i –Two output bits: Sum i, CarryOut i ( CarryIn i+1 = CarryOut i ) CarryIn... (0) (1) (0) (0) (0)... 0 0 1 0 1... 0 0 1 1 0... 0 (0) 1 (1) 0 (0) 1 (0) 1 + Inputs Outputs A: B: Sum : CarryOut

9 Full Adder’s Truth Table Sum A Symbol B CarryIn CarryOut + Definition A B CarryInCarryOutSum 00000 00101 01001 01110 10001 10110 11010 11111 CarryOut = (A’*B*CarryIn) + (A*B’*CarryIn) + (A*B*CarryIn’) + (A*B*CarryIn) = (B*CarryIn) + (A*CarryIn) + (A*B) Sum = (A’*B’*CarryIn) + (A’*B*CarryIn’) + (A*B’*CarryIn’) + (A*B*CarryIn)

10 Full Adder Circuit (1/2) 1.Construct the gates for Sum 2.Implement the gates for CarryOut 3.Connect all inputs with the same name

11 Full Adder Circuit (2/2)

12 One-Bit ALU Least significant bitOther bits

13 32-Bit ALU a0a0 a1a1 a 31 b0b0 b1b1 b 31... Result 0 Result 31 Result 1 BinvertOperation

14 Summary Building blocks: basic gates (AND, OR, NOT) Modular design and implementation –Gates have multiple inputs and one output –ALU works with 32-bit words (integers) –ALU implements a variety of operations in parallel => Construct first a 1-bit ALU –Mux chooses one of many different ALU operations –From the architecture’s instruction set, add the basic ALU operations necessary to implement that instruction –Two’s complement representation allows the use of the same hardware for both addition and subtraction

15 Anticipate the Weekend!!

16 Application to MIPS ALU MIPS ALU extensions Overflow detection Slt instruction Branch instructions Shift instructions Immediate instructions ALU performance –Performance vs. cost –Carry lookahead adder Implementation alternatives

17 Recall: Generic One-Bit ALU AND OR 00 – AND 01 – OR 10 – ADD First bit (LSB) Other bits 0 – ADD 1 – SUB Operations: AND, OR, ADD, SUB Control lines: 000 001 010 110

18 Slt Instruction Slt rd, rs, rt A A – B < 0 1.Perform subtraction using full adder 2.Check highest-order bit (sign bit) 3.Sign bit tells us whether A < B New input line ( Less ) goes directly to mux New control line (111) for slt Result for slt is not the output from ALU –Need a new 1-bit ALU for the most significant bit It has a new output line (Set) used only for slt (Overflow detection logic is also associated with this bit) 0000 0000 0000 0000 0000 0000 0000 000r rd: 1 if (rs < rt) 0 else

19 Slt Support First bit (LSB)Sign bit

20 Branch Instructions beq $t5, $t6, L –Use subtraction: (a-b) = 0 implies a = b –Add hardware to test if the result is 0 –OR all 32 results and invert the OR output Zero = (Result 1 + Result 2 +.. + Result 31 ) Consider A + B and A - B –Overflow if A = 0 ? B = 0 ?

21 Branch Support 1 (A = B) 0 otherwise

22 Shift instructions SLL, SRL, and SRA We need a data line for a shifter (L and R) However, shifters are much more easily implemented at the transistor level (outside the ALU) Barrel shifters x 3 x 2 x 1 x 0 x 2 x 1 x 0 00 x 3 x 2 x 1 Output, xOutput, x<<1Output, x>>1 Diagonal closed switch pattern controlled by the control unit

23 Immediate Instructions First input to ALU is the first register (rs) Second input –Data from register (rt) –Zero- or sing-extended immediate Add a mux at second input of ALU IR: Control Unit 0 1 Sign extend 1632 ALU Zero Overflow Result Registers Memory address rsrt

24 ALU Performance Is a 32-bit ALU as fast as a 1-bit ALU? –Can you see the ripple? Hardware executes in parallel Speed vs. Cost –Fewer sequential gates vs. number of gates Two extremes to do addition –Ripple carry and sum-of-products How could you get rid of the ripple –Two levels of logic c 1 = b 0 c 0 + a 0 c 0 + a 0 b 0 c 2 = b 1 c 1 + a 1 c 1 + a 1 b 1 c 2 = c 3 = b 2 c 2 + a 2 c 2 + a 2 b 2 c 3 = c 4 = b 3 c 3 + a 3 c 3 + a 3 b 3 c 4 =

25 Carry-Lookahead Adder (1/2) An approach in-between our two extremes Motivation: – If we didn't know the value of carry-in, what could we do? –When would we always generate a carry? g i = a i b i –When would we propagate the carry? p i = a i + b i Did we get rid of the ripple? c 1 = g 0 + p 0 c 0 c 2 = g 1 + p 1 c 1 c 2 = g 1 + p 1 g 0 + p 1 p 0 c 0 c 3 = g 2 + p 2 c 2 c 3 = c 4 = g 3 + p 3 c 3 c 4 =

26 Carry-Lookahead Adder (2/2) Can’t build a 16 bit adder this way... (too big) Could use ripple carry of 4-bit CLA adders Better: use the CLA principle again!

27 Second Level P and G

28 Ripple Carry vs. Carry Lookahead Assume each gate (AND or OR) takes the same time Total time = number of gates of longest path Consider 16-bit adders CarryOut signals c 16 and C 4 define the longest path –Ripple carry: 2 * 16 = 32 –Carry lookahead: 2 + 2 + 1 = 5 2 levels of logic in terms of P i and G i P i is specified in one level of logic (AND) using p i G i is specified in two levels of logic using p i and g i p i and g i are each one level of logic in terms of a i and b i A carry lookahead adder is six times faster

29 Implementation Alternatives The logic equation for the sum output can be expressed more simply with XOR gates Sum = a XOR b XOR CarryIn In some technologies, XOR is more efficient than two levels of AND and OR Processors are designed now in CMOS transistors (switches) CMOS ALU and barrel shifters have many fewer multiplexors than shown in our design However, the design principles are similar

30 Conclusions We can build an ALU to support the MIPS ISA –Key Idea: Use multiplexer to select ALU output –Subtraction uses two’s complement addition –Replicate 1-bit ALU to produce 32-bit ALU Important points about hardware –All of the gates in the ALU work in parallel –The speed of a gate is affected by the number of inputs –Speed of a circuit is affected by the number of gates in series (on the critical path or the deepest level of logic) Our primary focus: (conceptual) –Clever changes to organization can improve performance (similar to using better algorithms in software)

31 Enjoy the Weekend!!


Download ppt "CDA 3101 Fall 2013 Introduction to Computer Organization The Arithmetic Logic Unit (ALU) and MIPS ALU Support 20 September 2013."

Similar presentations


Ads by Google